Download Cloning, Sequencing, and Characterization of the Pradimicin

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Lac operon wikipedia , lookup

Expression vector wikipedia , lookup

Ribosomally synthesized and post-translationally modified peptides wikipedia , lookup

Ridge (biology) wikipedia , lookup

Genomic imprinting wikipedia , lookup

Copy-number variation wikipedia , lookup

Transcriptional regulation wikipedia , lookup

Real-time polymerase chain reaction wikipedia , lookup

Genetic engineering wikipedia , lookup

Gene expression wikipedia , lookup

Point mutation wikipedia , lookup

Gene therapy of the human retina wikipedia , lookup

Endogenous retrovirus wikipedia , lookup

RNA-Seq wikipedia , lookup

Gene therapy wikipedia , lookup

Promoter (genetics) wikipedia , lookup

Vectors in gene therapy wikipedia , lookup

Gene wikipedia , lookup

Gene desert wikipedia , lookup

Community fingerprinting wikipedia , lookup

Gene nomenclature wikipedia , lookup

Gene expression profiling wikipedia , lookup

Gene regulatory network wikipedia , lookup

Silencer (genetics) wikipedia , lookup

Biosynthesis of doxorubicin wikipedia , lookup

Artificial gene synthesis wikipedia , lookup

Transcript
. (2007), 17(5), 830–839
J. Microbiol. Biotechnol
Cloning, Sequencing, and Characterization of the Pradimicin Biosynthetic
Gene Cluster of Actinomadura hibisca P157-2
KIM, BYUNG CHUL1, JUNG MIN LEE1, JONG SEOG AHN2, AND BEOM SEOK KIM1*
1
Division of Biotechnology, College of Life Sciences & Biotechnology, Korea University, Seoul 136-713, Korea
2
Korea Research Institute of Bioscience and Biotechnology, Daejeon 305-806, Korea
Received: December 1, 2006
Accepted: January 14, 2007
Abstract Pradimicins are potent antifungal antibiotics having
an unusual dihydrobenzo[α]naphthacenequinone aglycone
substituted with D-alanine and sugars. Pradimicins are polyketide
antibiotics produced by Actinomadura hibisca P157-2. The
gene cluster involved in the biosynthesis of pradimicins was
cloned and sequenced. The pradimicin gene cluster was
localized to a 39-kb DNA segment and its involvement in the
biosynthesis of pradimicin was proven by gene inactivation of
prmA and prmB (ketosynthases α and β). The pradimicin gene
cluster consists of 28 open reading frames (ORFs), encoding
a type II polyketide synthase (PKS), the enzymes involved in
sugar biosynthesis and tailoring enzymes as well as two
resistance proteins. The deduced proteins showed strong
similarities to the previously validated gene clusters of
angucyclic polyketides such as rubromycin, griseorhodin, and
fredericamycin. From the pradimicin gene cluster, prmP3
encoding a component of the acetyl-CoA carboxylase complex
was disrupted. The production levels of pradimicins of the
resulting mutants decreased to 62% of the level produced by
the wild-type strain, which indicate that the acetyl-CoA
carboxylase gene would have a significant role in the
production of pradimicins through supplying the extender
unit precursor, malonyl-CoA.
Keywords: Actinomadura hibisca, pradimicin, type II
polyketide synthase, polyketide biosynthesis
Polyketides form a large group of biologically active natural
products that are synthesized by polyketide synthase
(PKS) complexes via a sequence of Claisen condensations
of acyl-CoA precursors [10]. Aromatic polyketides such
as daunorubicin, oxytetracycline, tetracenomycin, and
fredericamycin are synthesized by bacterial aromatic polyketide
*Corresponding author
Phone: 82-2-3290-3047; Fax: 82-2-921-1715;
E-mail: [email protected]
synthases, also termed type II PKSs [10]. Type II PKSs
minimally consist of ketoacyl synthase α (KSα), ketoacyl
synthase β (KSβ) (also known as chain-length factor), and
an acyl carrier protein [34]. The enzyme complexes catalyze
the biosynthesis of the polyketide backbone, which is
generated by condensation of an acetyl starter unit and
multiple malonyl-CoA extender units. The polyketide
backbone is later regiospecifically folded and cyclized by
aromatase and cyclase enzymes to form the characteristic
linear, angular, or discoid aromatic ring structures [11].
The structures are further modified by oxidation, reduction,
methylation, and glycosylation steps to yield the final
polyketide antibiotics. The aforementioned synthetic steps
provide a wide array of structural diversity to the polyketides.
Over the past decade, the gene clusters encoding many
polyketides have been cloned, sequenced [14], and identified,
including the angucyclic polyketides jadomycin [7],
rubromycin [22], landomycin [35], fredericamycin [34],
and oviedomycin [19]. The angucycline group of antibiotics
is a family of natural products of the type II polyketides.
The angucycline antibiotics share many features, including
a tetracyclic ben(a)anthracene aglycone system [24], but the
folding pattern of the nascent β-polyketide chains differs
from that observed in other well-analyzed type II polyketides
such as actinorhodin and tetracencomycin [25]. The isolation
of genes responsible for the production of an angucycline
can be expected to expand the number of genes currently
available for the combinatorial design of hybrid polyketides
[9, 17].
Pradimicins are potent antifungal antibiotics produced by
Actinomadura hibisca P157-2 [32] and A. verrucosospora
subsp. neohibisca [26]. They are angucycline antibiotics
having an unusual dihydrobenzo[α]naphthacenequinone
aglycon substituted with D-alanine and sugars, thomosamine
and xylose [32] (Fig. 1). The biosynthetic origin of pradimicins
was established by isotope labeling experiments [12],
revealing that pradimicin is a dodecaketide, which is most
PRADIMICIN BIOSYNTHETIC GENE CLUSTER
Fig. 1.
831
Proposed biosynthetic pathway for pradimicins.
likely derived from sequential condensations of 11 malonylCoA extender units and an acetyl-CoA starter unit (Fig. 1).
A previous genetic study claimed to have identified ORFs
that are putatively involved in pradimicin A biosynthesis
of A. hibisca [5]. However, no direct evidence for this
supposition was provided. Subsequently, ketosynthase
subunits that are essential for pradimicin biosynthesis
were identified from A. verrucosospora subsp. neohibisca
[4]. However, to date, no further results regarding the
pradimicin biosynthetic gene cluster, other than minimal
PKS genes, have been reported.
In the present study, the gene cluster for pradimicin
biosynthesis in A. hibisca P157-2 has been cloned,
sequenced, and confirmed by gene inactivation studies. The
sequence analysis has revealed good alignment between
the biosynthetic machinery of pradimicin and those of
other angucyclic antibiotics such as rubromycin, griseorhodin,
and fredericamycin. Ketosynthase genes in the pradimicin
biosynthetic gene cluster were disrupted to confirm the
involvement of the cloned gene cluster in pradimicin
biosynthesis. In addition, the acetyl-CoA carboxylase gene
in the pradimicin biosynthetic gene cluster was disrupted
to reveal that the gene product has a role in pradimicin
production, probably through the supply of the malonylCoA precursor.
MATERIALS AND METHODS
Bacterial Strains and Culture Conditions
All Escherichia coli strains used in this study were grown
following standard protocols [27]. E. coli DH5α was
employed for routine gene manipulation. The pGEM-T
Easy vector (Promega, Madison, WI, U.S.A.) was used for
cloning PCR fragments. The pradimicin producer A.
hibisca P157-2 (ATCC53557) [32] was maintained on a
agar medium composed of 20 g mannitol, 20 g soybean
meal, 20 g agar per 1 l of distilled water. For the broth
culture, A. hibisca was grown at 28 C in ISP2 medium (4 g
yeast extract, 10 g malt extract, 4 g dextrose, 1 l distilled
water) at pH 7.0. The pradimicin production medium
consisted of 30 g glucose, 30 g soybean meal, 5 g
pharmamedia, 1 g yeast extract, 3 g CaCO , and 1 l distilled
water. To produce pradimicin, a loop-full of A. hibisca
wild-type or mutant strains was inoculated into a baffled
1-l flask containing 100 ml of the production medium or
ISP2 medium and incubated at 28 C in the dark for 7 days
on a rotatory shaker at 170 rpm.
o
3
o
Analysis of Pradimicin Production
The liquid culture (5 l) of A. hibisca wild-type and mutant
strains was centrifuged at 8,000 rpm for 30 min to remove
cells. The supernatant was adjusted to pH 2.0 with 6 N
HCl and the supernatant was thoroughly stirred with a
mixture of n-butanol:methanol (100:1). The organic phase
was collected and concentrated in vacuo using a rotary
evaporator (Büchi, Flawil, Switzerland). The resulting residue
was suspended in 5 ml methanol and the crude extracts
were chromatographed on a Sephadex LH-20 column
(C26/100, GE Healthcare, Uppsala, Sweden). The Sephadex
LH-20 column was eluted with methanol at a 0.1 ml/min
flow rate. The major red-colored eluant was pooled and
analyzed by high performance liquid chromatography (HPLC)
using a Gilson HPLC system (Gilson, Middleton, WI, U.S.A.)
equipped with a semipreparative C reversed-phase column
(ODS-H80, 1×25 cm, 4 µm, YMC Co., Kyoto, Japan). The
column was eluted at a flow rate of 2 ml/min from 20%
methanol in water containing 0.1% formic acid to 95%
methanol in water containing 0.1% formic acid as a
linear gradient over a 50-min uninterrupted interval. The
chromatogram was monitored by absorbance at 210 nm.
The production rates of pradimicins A and B were quantified
18
832
KIM
et al.
by using standard compounds. Pradimicin standards were
prepared from the culture broth of wild-type A. hibisca by
purification using various chromatographic procedures, as
described previously [32]. The purified pradimicins were
confirmed by H-NMR spectroscopic analyses.
1
Construction of a Genomic Library and Screening
The total genomic DNA of A. hibisca P157-2 (grown
on ISP2 medium) was prepared according to the
acetyltrimethylammonium bromide procedure for the isolation
of genomic DNA [13]. A genomic DNA library was
constructed by using pCC2FOS vector as recommended
in the manufacturer’s protocol (Epicentre Biotechnologies,
Madison, WI, U.S.A.). Approximately 1,800 fosmid clones
were probed by using the DNA of a digoxigenin-labeled
ketosynthase gene, which had been amplified from the A.
hibisca genomic DNA using a primer set (KS-F: 5'AGC
CCA TCA AAG TCC/G ATG A/GTC GG-3'; KS-R: 5'CC
GGT GTT C/GAC C/GGC GTA GAA CCA GGC G-3')
[34]. All PCR amplifications were performed using the
Expanded High Fidelity PCR system according to the
manufacturer’s instruction (Roche, Basel, Switzerland).
Labeling was accomplished using the digoxigenin DNA
labeling kit (Roche Diagnosics Corp., Basel, Switzerland).
The colony and Southern hybridizations were performed
on Hybond N (Amersham Biosciences, Buckinghamshire,
U.K.) as recommended by the manufacturer.
Sequencing and Data Analysis
Fosmid DNAs were extracted from E. coli by alkaline lysis
[27]. Nucleotide sequences were determined by the dideoxy
chain termination method, using an automatic sequencer
(ABI Prism 3700 DNA sequencer, Applied Biosystems,
Foster City, CA, U.S.A.). The DNA sequences were
assembled using SeqMan II (DNAStar Inc., Maddison,
WI, U.S.A.). The assembled DNA and deduced protein
sequences were analyzed using the ORF finder provided
by DNAStar Inc., and confirmed by a BLAST database
search in the NCBI.
Targeted Disruption of Genes Involved in Pradimicin
Biosynthesis
The prmA and prmB (KSα and KSβ homologous genes) of
the pradimicin biosynthetic gene cluster was disrupted by
using the PCR-targeted Streptomyces gene replacement
method [6]. The aac(3)IV apramycin-resistant gene and
oriT were amplified from the pIJ773 disruption cassette [6]
using the primers PRMab-F 5'-CCCCGACGAGCAGGGGCTCGACGTCGGGGGCCGCACGTGattccggggatccgtc
gacc-3' and PRMab-R 5'-GCGGGTTGCCATCTCGTCCGTCCTCTCGGGTCGTTGTCAtgtaggctggagctgcttc-3'
(pIJ773 homologous sequence is in lowercase letters). The
resulting PCR-amplified DNA was used to replace the
prmAB genes in fosmid clone pBC3. The resulting pBC3m
was then transferred into A. hibisca using established
methods [6]. The double crossover mutants were selected
on mannitol soya flour agar medium (MS medium: 20 g
mannitol, 20 g soy flour, 20 g agar, 1 l distilled water)
amended with apramycin (50 µg/ml). The allelic replacement
of the prmAB genes in the BC3m mutant was confirmed by
Southern hybridization analysis, and PCR amplification
and sequencing.
For the gene disruption of orf1, the apramycin-resistant
gene and oriT was amplified using the primers PRM1-F 5'TTCGCCGTTGCAGCCACGCGACGACCGTACCGGG
AGATGattccggggatccgtcgacc-3' and PRM1-R 5'-TGACCACCGTGCTCAACGC CCTACGAACCCCCGCCGTCAtgtaggctggagctgcttc-3'. The resulting PCR-amplified DNA
was used to replace the orf1 gene in the fosmid clone
pBC4. The resulting pBC4m was then transferred into A.
hibisca using the established methods [6].
For the gene disruption of orf30, the apramycin-resistant
gene and oriT was amplified using the primers PRM30-F
5'-GGGACCAGGTCCCGGCGGGCCGGTTACGGTGACGGCATGattccggggatccgtcgacc-3' and PRM30-R 5'CTCCCCTGATCCCCGCCGCCCCGGCCAGAAGGCGGCTCAtgtaggctggagctgcttc-3'. The resulting PCR-amplified
DNA was used to replace the orf30 gene in fosmid clone
pBC3. The resulting pBC3-30m was then transferred into
A. hibisca using the established methods [6].
The gene prmP3 (acetyl-CoA carboxylase homologous
gene) was disrupted using the apramycin-resistant gene and
oriT, which was amplified using PRMp3-F 5'-GACCGTGACGACCGACACGACCGGCTCCGCCGCCGAGT
Gattccggggatccgtcgacc-3' and PRMp3-R 5'-ATGCCGTGTCCGGCGCCGCGGGGGATGTCTCGGCCATCAtgtaggctggagctgcttc-3'. The amplified DNA was used to replace
the prmP3 gene in fosmid clone pBC3. The resulting
pBC3-P3m was then transferred into A. hibisca using the
established methods [6]. The allelic replacement of the
orf1, orf30, and prmP3 genes in A. hibisca was confirmed
following the method described previously [6].
RESULTS AND DISCUSSION
Identification and Sequence Analysis of the Pradimicin
Gene Cluster
The pradimicin biosynthetic gene cluster of A. hibisca was
identified from a fosmid library of genomic DNA. A
fosmid clone harboring the pradimicin gene cluster (pBC3)
was identified by using the ketosynthase genes obtained by
PCR amplification as probes. The fosmid clone was used
as a probe to identify the other overlapping fosmid clone
pBC4. A region of 39 kb comprising two overlapping pBC3
and pBC4 fosmid clones was sequenced and analyzed for
the presence of open reading frames (ORFs) by use of the
SeqMan II software package and BLAST program to
PRADIMICIN BIOSYNTHETIC GENE CLUSTER
833
Gene cluster map of the pradimicin biosynthetic genes of A. hibisca P157-2. Proposed functions of individual ORFs are
summarized in Table 1.
Fig. 2.
compare with proteins in the databases. The sequence has
been deposited at the NCBI database under the accession
number EF151801. As the result of sequence analysis,
32 complete ORFs were identified (Fig. 2). The G+C
content of this DNA region was 73.2% as the typical
Streptomyces G+C bias content. The putative functions of
each ORF and their closest homologs are shown in Table 1.
Analysis of the gene products revealed strong similarities
to the previously validated polyketide gene clusters of
angucyclic antibiotics such as rubromycin, griseorhodin,
and fredericamycin. The DNA sequences of several genes
(prmA~prmK) of the prm gene cluster obtained in this
report were identical to the corresponding gene sequences
reported previously in A. hibisca [5]. PrmA, B, and C
products showed strong similarities (89, 82, and 71%
identity, respectively) with those of the ORF1, ORF2, and
ORF3 identified from A. verrucosospora [4].
Genes Putatively Involved in the Biosynthesis of the
Polyketide Moiety
Sequence analysis revealed the existence of the type II
minimal polyketide synthase homologous genes prmA,
prmB, and prmC, which encode KSα, KSβ, and acyl carrier
protein (ACP), respectively. The enzyme complexes would
catalyze the biosynthesis of the polyketide backbone,
which is later regiospecifically folded and cyclized by
cyclase enzymes to form the characteristic angular structures.
PrmA is highly similar to the KSα (RubA) found from
the rubromycin biosynthetic gene cluster [22]. All KSαs
contain active-site cysteine in a highly conserved sequence,
STGCTSGLD. The same sequence was found in PrmA
and contains Cys176 at the presumed active site. PrmB
showed a high level of similarity to the KSβ of chartreusin
PKS [36] and contains the characteristic Gln169 residue.
KSβ is similar to KSα but lacks the active site Cys, which is
replaced by Gln. The prmC gene is located downstream of
prmB and encodes an acyl carrier protein with the
characteristic conserved G(x)DS motif, which is involved
in the 4'-phosphopantetheine attachment for the posttranslational modification of the enzyme [16]. These
three enzymes comprise the minimal PKS and would be
responsible for the assembly of eleven malonyl-CoA extender
units and an initial acetyl-CoA starter unit through Claisen
condensations for the generation of the dodecaketide
(Fig. 1).
The pradimicin gene cluster contains three possible
cyclase genes, prmD, prmL, and prmK, whose products are
thought to be responsible for subsequent intramolecular
aldol reactions, and for producing the stepwise ring
closures that lead to the angular pentacyclic molecular
structure (Fig. 1). PrmD shows high sequence similarity to
the N-terminal region of RubF, GrhT, and TcmN. The
RubF, GrhT, and TcmN are bifunctional cyclases that
consist of two halves, being homologous to cyclases and
oxidoreductases/methyltransferases. They are responsible
for initial folding of the nascent linear polyketide intermediate
and subsequent cyclizations involved in the formation of
the aromatic rings [18, 22, 29]. The deduced protein PrmD
is similar to those cyclases and possibly has a similar role
in pradimicin biosynthesis, catalyzing the initial folding
of the linear dodecaketide intermediate. PrmK shows a
similarity to JadI (49% identity) of the jadomycin PKS.
The angucycline cyclase JadI is responsible for catalyzing
the cyclization of the fourth ring (the final ring of
jadomycin) [15]. PrmL shares sequence similarity to
FdmE (67% identity), GrhS (58% identity), and TcmJ
(55% identity). The roles of FdmE, GrhS, and TcmJ are
not clearly understood, but they seemingly enhance the
yields of cyclized products through stabilization of the
minimal PKS and cyclase protein complex [1, 18].
The product of prmG contains a short-chain alcohol
dehydrogenase domain and shares high sequence similarity
to RubG, a putative reductase gene. PrmG is probably
involved in the reduction of the keto group at C11 of the
nascent polyketide chain to a hydroxy group, a modification
required for the subsequent cyclization [22].
PrmF shares high sequence similarity to the C-terminal
region of TcmN (46% identity), which functions as an Omethyltransferase to catalyze the O-methylation of the
phenolic OH group of tetracenomycin [29]. By analogy,
PrmF presumably is the methyltransferase responsible for
the O-methylating OH group of ring A of pradimicin (Fig. 1).
The methyltransferase homologous to PrmF could also
be found in the fredericamycin (FdmN, 50% identity),
griseorhodin A (GrhL, 24% identity), and rubromycin
(RubO, 29% identity) biosynthetic pathways, all of which
834
KIM
Table 1.
Gene
prmP1
prmP2
prmP3
prmY
prmX
prmW
prmV
prmU
prmT
prmS
prmO
prmQ
prmR1
prmR2
prmN
prmM
prmL
prmA
prmB
prmC
prmD
prmE
prmF
prmG
prmH
prmI
prmJ
prmK
et al.
Proposed function of ORFs in the pradimicin biosynthetic gene cluster.
Size
Identity
(a.a.) Protein homologs (the accession numbers)
(%)
449
ZhuD (AAG30191) Streptomyces sp. R1128
63
AccC (YP713409) Frankia alni
64
154
AccB (YP713408) Frankia alni
70
ZhuE (AAG30192) Streptomyces sp. R1128
70
554
AccA (YP713407) Frankia alni
62
ZhuF (AAG30193) Streptomyces sp. R1128
58
331
RubN2 (CAI94677) S. achromogenes
68
Gdh (AAK83290) S. spinosa
65
356
SchS6 (CAF31364) Streptomyces SCC 2136
56
Gra-ORF16 (CAA09637) S. violaceoruber
56
402
CYP881 (CAE53712) S. peucetius
42
CYP (AAD28449) S. lavendulae
41
332
Gra-ORF29 (CAA09650) S. violaceoruber
54
SAV6641 (NP827817). S. avermitilis MA-4680
48
377
NmuI (YP410988) Nitrosospira multiformis
56
EffA (YP606036) Pseudomonas entomophila
54
340
TcmO (P39896) S. glaucescens
49
GrhL (AAM33664) Streptomyces JP95
45
438
RubG (CAC93713) Lachevalieria aerocolonigenes
38
AtG (ABC02800) A. melliaura
38
239
AknX2 (AAF73460) S. galilaeus
53
NbmE (AAM88354) S. narbonensis
55
436
GT family 28 (YP823260) Solibacter usitatus
41
GT family 28 (YP823259) Solibacter usitatus
39
348
CmrA (CAE17540) S. griseus
46
MtrA (CAK50797) S. argillaceus
44
260
SAV6712 (NP827888) S. avermitilis
47
TVN0936 (NP111455) Thermoplasma volcanium
28
621
FdmV (AAQ08933) S. griseus
48
RubR (AAM97368) S. collinus
47
154
Rxyl (YP645372) Rubrobacter xylanophilus
40
hypothetical protein (NP828636) S. avermitilis
38
67
148
FdmE (AAQ08915) S. griseus
58
GrhS (AAM33682) Streptomyces JP95
426
RubA (AAG03067) S. collinus
75
AspA (AAV48829) S. spiramyceticus
75
393
ChaB (CAH10162) S. chartreusis
62
OvmK (CAG14966) S. antibioticus
62
88
ZhuN (AAG30201) Streptomyces sp. R1128
55
RubC (AAG03069) S. collinus
41
154
RubF (AAG03070) S. collinus
67
GrhT (AAM33685) Streptomyces JP95
60
154
FdmM1 (AAQ08924) S. griseus
43
GrhM (AAM33665) Streptomyces JP95
37
342
TcmN (P16559) S. glaucescens
46
FdmN (AAQ08925) S. griseus
50
247
RubG (AAG03071) S. collinus
53
GrhO10 (AAM33668) Streptomyces JP95
52
114
GrhU (AAM33683) Streptomyces JP95
63
RubH (AAG03072) S. collinus
65
104
FdmQ (AAQ08928) S. griseus
46
RubT (AAM97374) S. collinus
41
412
CYP881 (CAE53712) S. peucetius
47
LnmA (AAN85514) S. atroolivaceus
47
115
JaDI (AAD37852) S. venezuelae
49
GrhQ (AAM33678) Streptomyces JP95
41
Proposed function
Biotin carboxylase
Biotin carboxyl carrier protein
Transcarboxylase
dNDP-D-glucose 4,6-dehydratase
dNDP-1-glucose synthase
Oxygenase
Oxygenase
Aminotransferase
Methyltransferase
Glycosyltransferase
Aminomethylase
Glycosyltransferase
ABC transport system ATP-binding protein
ABC-type multidrug transport system
Asparagine Synthetase
Unknown
Cyclase
KSα
KSβ
ACP
Cyclase
Oxygenase
Methyltransferase
Reductase
Oxygenase
Oxygenase
Oxygenase
Cyclase
PRADIMICIN BIOSYNTHETIC GENE CLUSTER
produce a polyketide moiety having a methoxyquinone
structure [18, 34].
At least three tailoring oxygenations appear to take
place during pradimicin biosynthesis. The oxygens at C6,
C8, and C20 do not originate from malonyl-CoA precursor
but are probably generated by monooxygenases. Three
monooxygenases (PrmJ, PrmW, and PrmV) were found in
the pradimicin gene cluster. PrmJ shares a high level of
similarity to LnmA, which is a cytochrome P450 hydroxylase
identified in the leinamycin gene cluster of S. atroolivaceus
[3]. PrmW is similar to the cytochrome P450 hydroxylases
found from the genome sequence of S. peucetius
ATCC27952 [23] and the mitomycin biosynthetic gene
cluster of S. lavenulae [21]. PrmV also contains the domain
of FMN-dependent monooxygenases, similar to the enzyme
that is obtained from the granaticin biosynthetic gene
cluster of S. vioaceoruber Tu22 [30].
PrmN shares a high level of sequence similarity to a
family of asparagine synthetases, such as FdmV and RubR.
FdmV has been proposed to be involved in transferring an
amino group from an amino acid to the fredericamycin
intermediate to afford the 2-pyridone moiety [34]. PrmN is
possibly responsible as well for transferring an amino acid
into the polyketide intermediate.
Three ORFs within the pradimicin gene cluster encode
proteins whose function cannot be predicted by sequence
comparison alone. PrmE, PrmH, and PrmI contain a
so-called “antibiotic monooxygenases domain” identified
from several Streptomyces sp. [34]. These genes (prmE,
prmH, prmI) have close homologs in the fredericamycin,
rubromycin, and griseorhodin gene cluster. Given the
structural and biosynthetic resemblance among pradimicin,
fredericamycin, rubromycin, and griseorhodin, these findings
suggest that such genes may play roles in the common
post-PKS oxidative tailoring steps [34].
Genes Putatively Involved in Sugar Biosynthesis
The genes prmQ, prmO, prmS, prmT, prmU, prmX, and
prmY appear to encode the enzymes related to sugar
biosynthesis and the subsequent glycosylation of pradimicin.
PrmY is most similar to dNDP-glucose 4,6-dehydratase from
the rhamnose biosynthesis pathway of Saccharopolyspora
spinosa [20]. PrmX resembles Gra-ORF16 (56% identity),
a dNDP-glucose synthase found from the granaticin
biosynthetic gene cluster of Streptomyces violaceoruber
[30]. These enzymes have been shown to catalyze the
first two conserved steps in the formation of dNDP-4keto-6-deoxy-glucose [20]. PrmU and PrmO resemble an
aminotransferase and aminomethylase, respectively, which
were found in a deoxyaminosugar biosynthetic gene cluster,
and are probably involved in production of the thomosamine
moiety. The presence of two glycosyltransferases in the
cluster is consistent with the fact that the pradimicin
aglycone is substituted by two different sugar moieties,
835
thomosamine and xylose (Fig. 1). PrmS shows a high level
of similarity to the glycosyltransferase (RubG) found in
the rubromycin biosynthetic gene cluster of Streptomyces
sp. The PrmQ most resembles a glycosyltransferase from
Solibacter usitatus.
Other Genes
PrmR1 and PrmR2 are most similar to the amino acid
sequences of a variety of known transporter genes. PrmR1
resembles the proteins of the daunorubicin-resistant ABC
transporter ATP-binding protein family, whereas PrmR2
shows a high degree of similarity to several ABC-type
multidrug transport systems. These genes are putatively
involved in pradimicin resistance, most probably via metabolite
export pumps powered by transmembrane electrochemical
gradients [34].
We found the presence of PrmP1, PrmP2, and PrmP3,
which are expected to be involved in the synthesis of a
malonyl-CoA extender unit. The PrmP1 and PrmP3 are
very similar to the α-chain (biotin carboxylase) and β-chain
(transcarboxylase) of the acetyl-CoA carboxylase complex.
PrmP2 resembles biotin carboxyl carrier proteins. PrmP1
is thought to act as a biotin carboxylase that incorporates
CO into a biotin carrier protein. PrmP3 would then
catalyze the transcarboxylation from biotin-CO to acetylCoA, thereby providing the malonyl-CoA precursor [8].
Other angucycline antibiotic gene clusters such as jadomyin
[33], landomycin [35], and simocyclinone [31] also
contain genes that are similar to prm1, prm2, and prm3.
2
2
Confirmation of the Pradimicin Gene Cluster by Gene
Inactivation
In order to confirm that the cloned genes were essential for
pradimicin biosynthesis, an in-frame deletion mutant of
A. hibisca was constructed (Fig. 3). The pBC3-based
plasmid pBC3m was created by replacing prmA and prmB
encoding KSα and Ksβ with the aac(3)IV apramycin-resistant
gene and oriT cassette by the PCR-targeted Streptomyces
gene replacement method [6]. The pBC3m was transferred
to wild-type A. hibisca by conjugation, and double crossover
mutants were selected (Fig. 3). HPLC analysis revealed
that the mutant completely lost its ability to produce
pradimicins, whereas wild-type strain produced 9.1 mg/l
of pradimicin A and 5.4 mg/l of pradimicin B (Fig. 3). This
result is a direct evidence for the involvement of the cloned
genes in the biosynthesis of pradimicin.
The pradimicin gene cluster boundaries were defined
by inactivation of selected genes (orf1 and orf30) residing
at the distal ends of the sequenced region. The gene
orf1 residing upstream of prmP1 encodes the protein that
resembles the 2,3,4,5-tetrahydropyridine-2,6-dicarboxylate
N-succinyltransferase (ZP00571119, 63% identity) of Frankia
sp. The gene of pBC4 was replaced with the apramycinresistant gene and oriT by the PCR-targeted Streptomyces
836
KIM
et al.
Confirmation of the cloned cluster to encode pradimicin biosynthesis by inactivation of prmA and prmB.
A. Isolation of prmAB deletion mutant BC3m and restriction maps of the A. hibisca wild-type and the mutant strain showing predicted fragment sizes upon
SphI digestion. B. Southern analysis of wild-type (lane 1) and BC3m (lane 2) genomic DNAs digested with SphI, using a 612-bp fragment of prmL as a
probe. C. HPLC analysis of the culture extract of the wild-type strain. Arrows indicate pradimicin A and pradimicin B. D. HPLC analysis of the culture
Fig. 3.
extract of the mutant strain BC3m.
gene replacement method. The resulting pBC4m was
transferred into A. hibisca by conjugation to give a double
crossover mutant. The mutant (BC4m) was selected in MS
medium amended with apramycin. The mutant still produced
pradimicin (8.9 mg/l pradimicin A and 5.8 mg/l pradimicin
B) at the similar level as wild-type A. hibisca, which
indicates that the deletion of the orf1 gene has little effect on
pradimicin production.
The gene orf30 locating downstream of prmK was also
replaced by the apramycin-resistant gene and oriT as
described previously. Orf30 is highly similar to a hypothetical
protein (NP733656, 42% identity) found from the genome
sequence of S. coelicolor. The mutant strain (BC3m30)
was fermented and examined for pradimicin production.
The effect of mutation on the level of pradimicin was
insignificant. The mutant produced 9.0 mg/l of pradimicin
A and 5.1 mg/l of pradimicin B. These results indicate that
orf1 and orf30 are either outside of the pradimicin gene
cluster or dispensable under the growth conditions used in
this study.
Effect of Gene Disruption of prmP3
The pradimicin polyketide moiety is generated by condensation
of an acetyl-CoA starter unit with eleven C extender units
supplied from malonyl-CoA. Malonyl-CoA is required
for chain elongation in all organisms that biosynthesize
fatty acids not only for the polyketide synthesis. Malonyl2
CoA is synthesized by carboxylation of acetyl-CoA. The
carboxyl group is introduced through two distinct reactions
catalyzed by a composite enzyme, acetyl-CoA carboxylase
(EC 6.4.1.2), which consists of a biotin carboxylase (BC),
a biotin carboxyl carrier protein (BCCP) with biotin as an
essential cofactor, and a transcarboxylase (TC). The initial
reaction, catalyzed by BC, uses bicarbonate to carboxylate
BCCP; the second reaction, catalyzed by TC, transfers
the carboxyl group from BCCP to acetyl-CoA, forming
malonyl-CoA [28]. The organization of genes encoding
the enzymes is variable in different organisms [28].
Most type II polyketide biosynthetic gene clusters do
not contain the enzymes because the malonyl-CoAs are
supplied by a primary metabolic pathway such as fatty acid
biosynthesis [2]. An interesting feature of the pradimicin
gene cluster is the presence of prmP1, prmP2, and prmP3,
which are homologous to BC, BCCP, and TC. Other
angucycline antibiotic gene clusters, such as jadomycin [33],
landomycin [35], and simocyclinone [31], also contain
genes encoding the acetyl-CoA carboxylase complex. The role
of these gene products would be in supplying a malonylCoA precursor to the pradimicin biosynthetic pathway
[33]. In order to know the effects of these genes, prmP3
encoding transcarboxylase was deleted from the gene
cluster as described previously (Fig. 4). The resulting
mutant (BC3-P3m) was selected from MS medium and the
production levels of pradimicins by the mutant were
PRADIMICIN BIOSYNTHETIC GENE CLUSTER
837
Inactivation of the prmP3 gene in the pradimicin biosynthetic gene cluster.
A. Isolation of prmP3 deletion mutant BC3-P3m and restriction maps of A. hibisca wild-type and the mutant strain showing predicted fragment sizes upon
NcoI and AgeI digestion. B. Southern analysis of wild-type (lane 1) and BC3-P3m (lane 2) genomic DNAs digested with NcoI and AgeI using an 825-bp
fragment of prmP1 as a probe. C. HPLC analysis of the culture extract of the wild-type strain. Arrows indicate pradimicin A and pradimicin B. D. HPLC
Fig. 4.
analysis of the culture extract of the mutant strain BC3-P3m. Arrows indicate pradimicin A and pradimicin B.
compared with those by wild-type A. hibisca. The production
levels of pradimicins by the mutant strain decreased to
62% of the level produced by the wild-type (Fig. 4). The
mutant produced 5.6 mg/l of pradimicin A and 3.4 mg/l of
pradimicin B. This result indicates that the prmP3 product
has a major role in the biosynthesis of the polyketide
intermediate that is processed to form pradimicins. However,
the pradimicin production was not completely inhibited in
the prmP3 deletion mutant, which may be due to metabolic
leakage from the primary biosynthetic pathways. However,
it is apparent that the prmP3 homolog in the primary
metabolic pathway is unable to fully complement the
defect in the prmP3 deletion mutant. In a previous report
on jadomycin biosynthesis, the deletion of the acetyl-CoA
carboxylase gene jadJ resulted in a 85% reduction of
the jadomycin production [7]. These results indicate that
the acetyl-CoA carboxylase gene prmP3 would have a
significant role in the production of pradimicin by supplying
a precursor pool of malonyl-CoA.
We have confirmed that the cloned gene cluster was
responsible for the production of pradimicin by using a
mutant in which the gene encoding the PKS (KSα and KSβ)
was disrupted by in-frame deletion. Pradimicin is an unusual
dodecaketide angucycline antibiotic compound. The type
II polyketide synthases producing the angucycline structures
larger than dodecaketide are the griseorhodin, fredericamycin,
and rubromycin polyketide synthases. Although the angucycline
structures of these compounds are further processed to
generate a chiral spiro-carbon center [34], the fact that the
pradimicin biosynthetic machinery shares similarities with
the oxygenases and cyclases known for the griseorhodin,
fredericamycin, or rubromycin biosynthetic machinery supports
the suggestion of a common mechanism for forming the
angucycline ring moiety. The availability of the pradimicin
biosynthetic genes described in this report would provide
new tools to study the unusual biochemistry of angucycline
biosynthesis, and to refine the understanding of the cyclization
mechanism, and also provide the tools for combinatorial
biosynthetic studies to generate novel angucycline antibiotics.
Acknowledgments
The plasmids and the strains used for the PCR-targeted
Streptomyces gene replacement were kindly provided by
Dr. Bertolt Gust in Plant Bioscience Limited, Norwich,
England. This work was supported by the BioGreen21
Program, Rural Development Administration, Korea. J.S.A.
is supported in part by the 21C Frontier Microbial Genomics
and Application Center Program. NMR spectroscopic
analyses were performed with the help from the Korea
Basic Science Institute, Seoul.
838
KIM
et al.
REFERENCES
1. Bao, W., E. Wendt-Pienkowski, and C. R. Hutchinson. 1998.
Reconstitution of the iterative type II polyketide synthase for
tetracenomycin F2 biosynthesis. Biochemistry
81328138.
2. Chater, K. F. and M. J. Bibb. 1997. Regulation of bacterial
antibiotic production. In Kleinkauf, H. and H. von Dohren
(eds.). Biotechnology, vol. 7, Products of Secondary
Metabolism. VCH, Weinheim.
3. Cheng, Y. Q., G. L. Tang, and B. Shen. 2002. Identification
and localization of the gene cluster encoding biosynthesis
of the antitumor macrolactam leinamycin in Streptomyces
atroolivaceus S-140. J. Bacteriol.
7013-7024.
4. Dairi, T., Y. Hamano, T. Furumai, and T. Oki. 1999.
Development of a self-cloning system for Actinomadura
verrucosospora and identification of polyketide synthase
genes essential for production of the angucyclic antibiotic
pradimicin. Appl. Environ. Microbiol. 2703-2709.
5. Dairi, T., Y. Hamano, Y. Igarashi, T. Furumai, and T. Oki.
1997. Cloning and nucleotide sequence of the putative
polyketide synthase genes for pradimicin biosynthesis from
Actinomadura hibisca. Biosci. Biotechnol. Biochem.
1445-1453.
6. Gust, B., G. L. Challis, K. Fowler, T. Kieser, and K. F.
Chater. 2003. PCR-targeted Streptomyces gene replacement
identifies a protein domain needed for biosynthesis of the
sesquiterpene soil odor geosmin. Proc. Natl. Acad. Sci. USA
1541-1546.
7. Han, L., K. Yang, K. Kulowski, E. Wendt-Pienkowski, C. R.
Hutchinson, and L. C. Vining. 2000. An acyl-coenzyme A
carboxylase encoding gene associated with jadomycin
biosynthesis in Streptomyces venezuelae ISP5230. Microbiology
903-910.
8. Hill, A. M. 2006. The biosynthesis, molecular genetics and
enzymology of the polyketide-derived metabolites. Nat.
Prod. Rep. 256-320.
9. Hong, J., C. Y. Choi, and Y. J. Yoon. 2003. Premature release
of polyketide intermediates by hybrid polyketide synthase in
Amycolatopsis mediterranei S699. J. Microbiol. Biotechnol.
613-619.
10. Hopwood, D. A. 1997. Genetic contributions to understanding
polyketide synthases. Chem. Rev. 2465-2497.
11. Jakobi, K. and C. Hertweck. 2004. A gene cluster encoding
resistomycin biosynthesis in Streptomyces resistomycificus;
exploring polyketide cyclization beyond linear and angucyclic
patterns. J. Am. Chem. Soc.
2298-2299.
12. Kakushima, M., Y. Sawada, M. Nishio, T. Tsuno, and T. Oki.
1989. Biosynthesis of pradimicin A. J. Org. Chem.
2536-2539.
13. Kieser, T., M. J. Bibb, M. J. Buttner, K. F. Chater, and D. A.
Hopwood. 2000. Practical Streptomyces Genetics. The John
Innes Foundation, Norwich, United Kingdom.
14. Kim, D., Y. K. Park, J. S. Lee, J. F. Kim, H. Jeong,
B. S. Kim, and C. H. Lee. 2006. Analysis of a prodigiosin
biosynthetic gene cluster from the marine bacterium Hahella
chejuensis KCTC2396. J. Microbiol. Biotechnol.
19121918.
37:
184:
65:
61:
100:
146:
23:
13:
97:
126:
54:
16:
15. Kulowski, K., E. Wendt-Pienkowski, L. Han, K. Yang,
L. C. Vining, and C. R. Hutchinson. 1999. Functional
characterization of the jadI gene as a cyclase forming
angucyclinones. J. Am. Chem. Soc.
1786-1794.
16. Lambalot, R. H., A. M. Gehring, R. S. Flugel, P. Zuber,
M. LaCelle, M. A. Marahiel, R. Reid, C. Khosla, and
C. T. Walsh. 1996. A new enzyme superfamily - the
phosphopantetheinyl transferases. Chem. Biol. 923-936.
17. Lee, S., J. Park, S. Park, J. S. Ahn, C. Choi, and Y. J. Yoon.
2006. Hydroxylation of indole by PikC cytochrome P450
from Streptomyces venezuelae and engineering its catalytic
activity by site-directed mutagenesis. J. Microbiol. Biotechnol.
974-978.
18. Li, A. and J. Piel. 2002. A gene cluster from a marine
Streptomyces encoding the biosynthesis of the aromatic
spiroketal polyketide griseorhodin A. Chem. Biol. 10171026.
19. Lombo, F., A. F. Brana, J. A. Salas, and C. Mendez. 2004.
Genetic organization of the biosynthetic gene cluster for the
antitumor angucycline oviedomycin in Streptomyces antibioticus
ATCC 11891. Chembiochem 1181-1187.
20. Madduri, K., C. Waldron, and D. J. Merlo. 2001. Rhamnose
biosynthesis pathway supplies precursors for primary and
secondary metabolism in Saccharopolyspora spinosa. J.
Bacteriol.
5632-5638.
21. Mao, Y., M. Varoglu, and D. H. Sherman. 1999. Molecular
characterization and analysis of the biosynthetic gene cluster
for the antitumor antibiotic mitomycin C from Streptomyces
lavendulae NRRL 2564. Chem. Biol. 251-263.
22. Metsa-Ketela, M., K. Ylihonko, and P. Mantsala. 2004.
Partial activation of a silent angucycline-type gene cluster
from a rubromycin beta producing Streptomyces sp. PGA64.
J. Antibiot. 502-510.
23. Parajuli, N., D. B. Basnet, C. H. Lee, J. K. Sohng, and K.
Liou. 2004. Genome analyses of Streptomyces peucetius
ATCC 27952 for the identification and comparison of
cytochrome P450 complement with other Streptomyces.
Arch. Biochem. Biophys.
233-241.
24. Rohr, J. 2000. Bioorganic Chemistry, Deoxysugars, Polyketides
121:
3:
16:
9:
5:
183:
6:
57:
425:
and Related Classes: Synthesis, Biosynthesis, Enzymes.
Springer, Germany.
25. Rohr, J. and K. Thiericke. 1992. Angucycline group antibiotics.
Nat. Prod. Rep. 103-137.
26. Saitoh, K., Y. Sawada, K. Tomita, T. Tsuno, M. Hatori,
and T. Oki. 1993. Pradimicins L and FL: New pradimicin
congeners from Actinomadura verrucosospora subsp.
neohibisca. J. Antibiot. 387-397.
27. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular
Cloning: A Laboratory Manual, 2nd Ed. Cold Spring Harbor
Laboratory Press, Cold Spring Harbor, NY.
28. Samols, D., C. G. Thornton, V. L. Murtif, G. K. Kumar, F. C.
Haase, and H. G. Wood. 1988. Evolutionary conservation
among biotin enzymes. J. Biol. Chem.
6461-6464.
29. Shen, B., R. G. Summers, E. Wendt-Pienkowski, and C. R.
Hutchinson. 1995. The Streptomyces glaucescens tcmKL
polyketide synthase and tcmNM polyketide cyclase genes
govern the size and shape of aromatic polyketides. J. Am.
Chem. Soc.
6811-6821.
9:
46:
263:
117:
PRADIMICIN BIOSYNTHETIC GENE CLUSTER
30. Sherman, D. H., F. Malpartida, M. J. Bibb, H. M. Kieser, M.
J. Bibb, and D. A. Hopwood. 1989. Structure and deduced
function of the granaticin-producing polyketide synthase
gene cluster of Streptomyces violaceoruber Tu22. EMBO J.
2717-2725.
31. Trefzer, A., S. Pelzer, J. Schimana, S. Stockert, C. Bihlmaier,
H. P. Fiedler, K. Welzel, A. Vente, and A. Bechthold.
2002. Biosynthetic gene cluster of simocyclinone, a natural
multihybrid antibiotic. Antimicrob. Agents Chemother.
1174-1182.
32. Tsunakawa, M., M. Nishio, H. Ohkuma, T. Tsuno, M.
Konishi, T. Naito, T. Oki, and H. Kawaguchi. 1989. The
structure of pradimicins A, B, and C: A novel family of
antifungal antibiotics. J. Org. Chem. 2532-2536.
33. Wang, L., J. McVey, and L. C. Vining. 2001. Cloning and
functional analysis of a phosphopantetheinyl transferase
superfamily gene associated with jadomycin biosynthesis in
8:
46:
54:
839
Streptomyces venezuelae ISP5230. Microbiology
15351545.
34. Wendt-Pienkowski, E., Y. Huang, J. Zhang, B. Li, H. Jiang,
H. Kwon, C. R. Hutchinson, and B. Shen. 2005. Cloning,
sequencing, analysis, and heterologous expression of the
fredericamycin biosynthetic gene cluster from Streptomyces
griseus. J. Am. Chem. Soc.
16442-16452.
35. Westrich, L., S. Domann, B. Faust, D. Bedford, D. A. Hopwood,
and A. Bechthold. 1999. Cloning and characterization of a
gene cluster from Streptomyces cyanogenus S136 probably
involved in landomycin biosynthesis. FEMS Microbiol. Lett.
381-387.
36. Xu, Z., K. Jakobi, K. Welzel, and C. Hertweck. 2005.
Biosynthesis of the antitumor agent chartreusin involves the
oxidative rearrangement of an anthracyclic polyketide. Chem.
Biol. 579-588.
147:
127:
170:
12: